Heltec 350A relay BMS (48V / 16S)

[fafrd]

I built two 24V / 8S LiFePO4 batteries with Heltec BMSes and have been happy with them.

I’m now planning my first 16S / 48V LiFePO4 battery and perusing Heltec’s BMS offerings, I see they now offer a 350A relay-based BMS: https://heltec-bms.com/product/350a...s-8s-10s-12s-13s-16s-17s-20s-24s-28s-32s-35s/

I’m interested in whether any members have experience with this Heltec relay-based BMS or other relay-based BMSes.

Also interested in any general pros and cons of rely-based BMSes versus the typical solid-state variety.

Lastly, if anyone has a reasonably-priced 48V / 16S BMS to recommend, appreciate any advice.

 

[BentleyJ]

You have asked about one of my favorite subjects. For larger, high amperage battery systems I much prefer a big beefy mechanical contactor. The problem with them is the 24/7 parasitic load required to keep the coil energized and thus the contacts closed. I found that the simple low cost golf cart style contactor coils draw from 120mA, 200A contact rating to over 300mA for larger 400A to 600A devices. There are more expensive contactors with economizers and dual coil set ups where the holding current is lower than the closing current but its surprising how much power is going to waste on a monthly basis when you actually do the math.

The other issue with a mechanical contactor is their bi-directional nature so in many cases there is no way to disable charging without disabling discharging as well. The obvious solution to this issue is to have the BMS sense when conditions have become favorable to reclose the contactor. Does the Heltec do this? I believe there is a QUCC brand that also uses a contactor.

Ultimately the above led to the idea of using magnetic latching contactors or a remote trip breaker. Both of which are not natively compatible with a typical BMS ST output relay. Batrium seems to have a built in pulse capability.
Latching contactors come in both single and dual coil designs, each requiring slightly different control signals and usually require a pulse of current to operate but can be damaged by continuous current.
Likewise some remote trip breakers require a pulse to trip but sustained current will damage the coil. There are other RT breaker brands that cut power to the coil at the same time the breaker trips.

I have previously posted information here and on YT regarding several PCB controllers I've developed for latching contactors and RT breakers.

At this time I'm using a 400A DuraKool latching contactor on a DIY battery with the control board I designed connected to a ZEVA BMS which uses photo relays to control charge/discharge cut out.

 

[Rocketman]

I used a Batrium Core BMS for my MotorHome 12v 544ah battery. I used this Bms because I wanted possible very high current availability. For the last 2+ years it was worked very well.

Two years ago my choices that I knew about were: Batrium, Rec-Bms, and Orion. All of these are more expensive bms’s but from what I understand they are all more robust than FET based bms’s (or at least that’s what it seemed like 2 years ago).

For my relay I used an Tyco EV200AAANA- this is 9v to 36v on the coils, they also have a 48v too.

Good Luck

 

[fafrd]

You have asked about one of my favorite subjects. For larger, high amperage battery systems I much prefer a big beefy mechanical contactor. The problem with them is the 24/7 parasitic load required to keep the coil energized and thus the contacts closed.

Since the contactor is closed 99.99999% or the time (always unless there is a problem / failure), why not use a normally-closed contactor where AC energy gets expended only to keep the contractor open to protect the battery?

I found that the simple low cost golf cart style contactor coils draw from 120mA, 200A contact rating to over 300mA for larger 400A to 600A devices. There are more expensive contactors with economizers and dual coil set ups where the holding current is lower than the closing current but its surprising how much power is going to waste on a monthly basis when you actually do the math.

Understand, thanks. Would still like to know why normally closed would not be a solution…

The other issue with a mechanical contactor is their bi-directional nature so in many cases there is no way to disable charging without disabling discharging as well.

Why is this a problem? I’m only expecting the BMS to cut out when there is a failure in the system and I don’t mind the whole system staying shut down until I manually reset it.

I have no interest in relying on the BMS to shut down power generation once the battery is drained - learned the hard way how damaging that can be to an inverter.

So I think I’m OK having both charge and discharge cut off together…

The obvious solution to this issue is to have the BMS sense when conditions have become favorable to reclose the contactor. Does the Heltec do this? I believe there is a QUCC brand that also uses a contactor.

I’ll check, but again, don’t believe it is critical for my use case. I intend BMS protection to only kick in when there has been a major issue that requires understanding / resolution before the system starts running again…

Ultimately the above led to the idea of using magnetic latching contactors or a remote trip breaker. Both of which are not natively compatible with a typical BMS ST output relay. Batrium seems to have a built in pulse capability.
Latching contactors come in both single and dual coil designs, each requiring slightly different control signals and usually require a pulse of current to operate but can be damaged by continuous current.
Likewise some remote trip breakers require a pulse to trip but sustained current will damage the coil. There are other RT breaker brands that cut power the coil at the same time the breaker trips.

I have previously posted information here and on YT regarding several PCB controllers I've developed for latching contactors and RT breakers.

At this time I'm using a 400A DuraKool latching contactor on a DIY battery with the control board I designed connected to a ZEVA BMS which uses photo relays to control charge/discharge cut out.

I looked into using contactors once but they were pricey. Part of what attracted me to that Heltec offering was little/no premium over conventional FET-based options.

 

[fafrd]

I used a Batrium Core BMS for my MotorHome 12v 544ah battery. I used this Bms because I wanted possible very high current availability. For the last 2+ years it was worked very well.

Two years ago my choices that I knew about were: Batrium, Rec-Bms, and Orion. All of these are more expensive bms’s but from what I understand they are all more robust than FET based bms’s (or at least that’s what it seemed like 2 years ago).

For my relay I used an Tyco EV200AAANA- this is 9v to 36v on the coils, they also have a 48v too.

Good Luck

Thanks. I considered the Batrium but it plus Contactor were to pricey for me. That is what caught my eye about this Heltec option - little to know price premium over their FET-based BMSes.

One of the appealing aspects of relay-based for me is that heat generation in the BMS itself should be much lower than for FET-based.

 

[BentleyJ]

Since the contactor is closed 99.99999% or the time (always unless there is a problem / failure), why not use a normally-closed contactor where AC energy gets expended only to keep the contractor open to protect the battery?

Understand, thanks. Would still like to know why normally closed would not be a solution…

Yes, absolutely a NC contactor would work, I didn't mention it because I believe it has a few negatives.
1) If a critical battery condition occurs the point is to disconnect the load. A NC contactor requires power to maintain a disconnected state. You are then relying on a component that has a known problem but has to function properly to keep itself off-line. Poor engineering.
2) It would require the BMS to function properly and maintain a sustained "active" state this is also poor engineering practice for fail safe devices. Passive safegurards are better. The controller PCB's I developed have on-board capacitors that will open a latching contactor for any reason the control signal is cut. Could a BMS fail in such a way that the relay control signal persisted even though the BMS otherwise was not working properly? Sure but much less likely.
3) NC contactors are less common and not always stocked. Although this is definitely true for magnetic latching contactors as well.

I’m only expecting the BMS to cut out when there is a failure in the system and I don’t mind the whole system staying shut down until I manually reset it.
I have no interest in relying on the BMS to shut down power generation once the battery is drained - learned the hard way how damaging that can be to an inverter.

So I think I’m OK having both charge and discharge cut off together…

Very true, and this logic is exactly why I would actually choose the remote trip breaker option because under normal circumstances the BMS should not trip. Thus manual intervention is probably needed anyway and its not such a big concern that the system needs to automatically restart itself.

I’ll check, but again, don’t believe it is critical for my use case. I intend BMS protection to only kick in when there has been a major issue that requires understanding / resolution before the system starts running again…

Absolutely, I agree. In your OP you asked for a discussion of General Pros and Cons. This is what we are doing. I wasn't advocating for one thing over another.

I looked into using contactors once but they were pricey. Part of what attracted me to that Heltec offering was little/no premium over conventional FET-based options.

True but there is a trade off. If a system is designed to have a certain kWh rating to carry the loads for a specified period of time, parasitic losses have to be accounted for and the system has to be up-sized accordingly which is not free.
Likewise, value priced All-in-One inverters have lower up front costs to the buyer but then they also waste 2 to 3 times as much energy for stand-by and operating losses as compared to higher priced brands. As I mentioned, when you do the math wasted energy adds up fast when its a 24/7 issue.

 

[fafrd]

Yes, absolutely a NC contactor would work, I didn't mention it because I believe it has a few negatives.
1) If a critical battery condition occurs the point is to disconnect the load. A NC contactor requires power to maintain a disconnected state. You are then relying on a component that has a known problem but has to function properly to keep itself off-line. Poor engineering.

Fair point, though even NC contact its can fail in a stuck closed condition, no?

Backup protection, etc is a whole ‘nuther topic.

Looking at the Heltec offering more closely, the relay they are using appears to be dual-coil design (2 pairs of control leads and a add-on rectangle) so hopefully NO power consumption is more modest.

They claim 80uA self consumption reducing to only 60uA after disconnect, but I suspect this is not including the consumption of the relay.

Also, for purposes of scale, my use case is powering 300W of load (fridges) 24/7, so I could accept as much as 30W of coil power but would be happier with a fraction of that.

In any case, you’ve confirmed ON consumption of the relay itself is perhaps the most significant drawback of the relay approach to BMS, so thanks.

2) It would require the BMS to function properly and maintain a sustained "active" state this is also poor engineering practice for fail safe devices. Passive safegurards are better. The controller PCB's I developed have on-board capacitors that will open a latching contactor for any reason the control signal is cut. Could a BMS fail in such a way that the relay control signal persisted even though the BMS otherwise was not working properly? Sure but much less likely.

Again, whole ‘nuther’ discussion (probably best for a separate thread). I am planning to eventually have a PLC monitoring the entire setup and will probably have a second relay/contactor to shut everything down.

The possibility of two NC contactors failing to open from the same failure event is risk I’m comfortable with…

3) NC contactors are less common and not always stocked. Although this is definitely true for magnetic latching contactors as well.

Good point. If I start with a NO contactor, that design can probably be switched to NC using an intermediary relay or signal inverter.

Very true, and this logic is exactly why I would actually choose the remote trip breaker option because under normal circumstances the BMS should not trip. Thus manual intervention is probably needed anyway and its not such a big concern that the system needs to automatically restart itself.

Agreed.

Absolutely, I agree. In your OP you asked for a discussion of General Pros and Cons. This is what we are doing. I wasn't advocating for one thing over another.

Understood and appreciated.

True but there is a trade off. If a system is designed to have a certain kWh rating to carry the loads for a specified period of time, parasitic losses have to be accounted for and the system has to be up-sized accordingly which is not free.
Likewise, value priced All-in-One inverters have lower up front costs to the buyer but then they also waste 2 to 3 times as much energy for stand-by and operating losses as compared to higher priced brands. As I mentioned, when you do the math wasted energy adds up fast when it’s a 24/7 issue.

I’m in an unusual position of having excess energy to waste (within limits). Either I’ll be throttling my solar power because battery is fully-charged and export limit has been reached or I can burn a bit of that power that otherwise would be getting dumped.

More worried about damaging components than I am about achieving highest possible efficiency.

But yeah, +10% to power the relay is about the most that would make sense.

I’ll reach out to Heltec to see what they can tell me about the relay they are using…

 

[SammyMc]

They have a new 500A version

 

[SammyMc]

anyone has different experience ?